The goals of this proposal are to identify and characterize the DNA sequences and nuclear protein factors responsible for the increased production of erythropoietin (Epo) in response to physiologic stimuli and for the tissue-specific expression of the human Epo gene. Studies of the stimulated production of Epo will employ the Hep3B hepatoma cell line which markedly increases Epo synthesis in response to hypoxia. DNA fragments in and near the Epo gene will be tested for regulatory activity by transfection assays in normoxic and hypoxic Hep3B cells. Active DNA fragments will probe nuclear protein extracts from Hep3B cells to identify the DNA-protein interactions responsible for enhanced Epo production. In vivo footprint studies will be performed on normoxic and hypoxic Hep3B cells to define the DNA-protein interactions present in living cells. Transgenic mice containing the human Epo gene will be made anemic or hypoxic to stimulate Epo expression. DNasel analysis will be performed on tissues from these mice to locate regulatory elements involved in Epo gene response to hypoxia in renal and hepatic cells. An expression cDNA library will be constructed from hypoxic Hep3B cells and screened with oligonucleotide probes that contain binding sites for nuclear proteins involved in the response to hypoxia in order to isolate the genes encoding these proteins. In order to define the regulatory elements that direct Epo gene expression to renal and hepatic cells, transgenic mice will be constructed with different genomic human Epo gene fragments and assayed for tissue distribution of Epo MRNA. Transgenic lines with appropriate tissue- specific expression will be examined for Dnasel hypersensitive sites in kidney, liver and other cells. The regulatory role of DNA fragments. Active DNA fragments will be used as probes in DNA-protein binding assays using nuclear protein extracts derived from cells or cell lines which make Epo. Additionally, in vivo footprint studies of these DNA fragments will be performed on liver and kidney cells from transgenic mice. To facilitate the study of tissue-specific expression, two approaches will be used in an attempt to derive a renal cell line with regulated Epo expression: first, kidney cells from transgenic mice will be immortalized with a temperature- sensitive Simian virus 40(SV40) which permits expression of a differentiated cellular phenotype in cells grown at an appropriate temperature; second, transgenic mice will be created with a construct in which the SV40 large T antigen gene is linked to sequences which direct Epo gene expression to renal cells. Protein factors which direct Epo gene expression to renal or hepatic cells will be cloned from CDNA libraries derived from anemic mouse kidney and liver cells.